简体繁体 English

MySQL计算总行数

[英]MySQL count total number of rows

原文 2010-03-03 02:09:12 2 2 mysql/ count

I have a complicated MySQL query which takes a lot of time, selecting from a table with more than 150k rows and several JOINS and subqueries. 我有一个复杂的MySQL查询，需要花费很多时间，从一个行超过150k的表和几个JOINS和子查询中进行选择。 I am limiting the amount of results, but I need to know the total amount of rows. 我限制了结果的数量，但我需要知道总行数。

Is there a way not to repeat the query with COUNT(*) as the field to get them? 有没有办法不用COUNT(*)重复查询作为获取它们的字段？ (repeating the query almost doubles the amount of time the script takes to complete) （重复查询几乎使脚本完成所需的时间翻倍）

2 个解决方案

MySQL has a clause called SQL_CALC_FOUND_ROWS which you put after SELECT and before any field names. MySQL有一个名为SQL_CALC_FOUND_ROWS的子句，它放在SELECT之后和任何字段名之前。 Then you call a second query which is just SELECT FOUND_ROWS() AS rows and it gives you how many rows the previous query found. 然后调用第二个查询，它只是SELECT FOUND_ROWS() AS rows ，它给出了上一个查询找到的行数。

It may actually be faster to make the two queries (so you may just want to do some quick testing): 实际上两个查询可能会更快（所以你可能只想做一些快速测试）：

http://www.mysqlperformanceblog.com/2007/08/28/to-sql_calc_found_rows-or-not-to-sql_calc_found_rows/ http://www.mysqlperformanceblog.com/2007/08/28/to-sql_calc_found_rows-or-not-to-sql_calc_found_rows/

In any case, try to make sure that the COUNT(*) can use the index (it will say "using index" in the Extra column if you use EXPLAIN query) 在任何情况下，尝试确保COUNT（*）可以使用索引（如果使用EXPLAIN查询，它将在Extra列中说“using index”）

You could also consider caching the COUNT(*) result if it doesn't have to be exact (as you're limiting your result). 您也可以考虑缓存COUNT（*）结果，如果它不必是精确的（因为您限制了结果）。